
    <hR6                     8    S SK Jr  S SKJr   " S S\5      rS/rg)   )PretrainedConfig)rope_config_validationc                      ^  \ rS rSrSrSrS/r0 SS_SS_SS_S	S
_SS
_SS_SS_SS_SS_SS_SS_SS_SS
_SS_SS_SS_rS/S/4SS/S/4S/S/4S.r                          S!U 4S jjr	S r
U =r$ )"
DogeConfig   a0  
This is the configuration class to store the configuration of a [`DogeModel`]. It is used to instantiate an Doge
model according to the specified arguments, defining the model architecture like [SmallDoge/Doge-320M](https://huggingface.co/SmallDoge/Doge-320M).

Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.

Args:
    vocab_size (`int`, *optional*, defaults to 32768):
        Vocabulary size of the Doge2 model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`DogeModel`]
    hidden_size (`int`, *optional*, defaults to 1024):
        Dimension of the hidden representations.
    intermediate_size (`int`, *optional*, defaults to 2048):
        Dimension of the MLP representations.
    num_hidden_layers (`int`, *optional*, defaults to 32):
        Number of hidden layers in the Transformer decoder.
    hidden_dropout (`float`, *optional*, defaults to 0.0):
        Dropout probability for each sequence transformation and state transformation module.
    hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
        The non-linear activation function (function or string) in the decoder.
    initializer_range (`float`, *optional*, defaults to 0.02):
        The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
    rms_norm_eps (`float`, *optional*, defaults to 1e-06):
        The epsilon used by the rms normalization layers.
    use_cache (`bool`, *optional*, defaults to `True`):
        Whether or not the model should return the last key/values attentions (not used by all models). Only
        relevant if `config.is_decoder=True`.
    tie_word_embeddings (`bool`, *optional*, defaults to `False`):
        Whether the model's input and output word embeddings should be tied.
    max_position_embeddings (`int`, *optional*, defaults to 2048):
        The maximum sequence length that this model might ever be used with.
    rope_theta (`float`, *optional*, defaults to 10000.0):
        The base period of the RoPE embeddings.
    rope_scaling (`Dict`, *optional*):
        Dictionary containing the scaling configuration for the RoPE embeddings.
        NOTE: if you apply new rope type and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value accordingly.
        Doge family of small models use `{ 'rope_type': 'dynamic', 'factor': 4.0, 'original_max_position_embeddings': 2048 }` as the default value.
        Expected contents:
            `rope_type` (`str`):
                The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope', 'llama3'], with 'default' being the original RoPE implementation.
            `factor` (`float`, *optional*):
                Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings.
                In most scaling types, a `factor` of x will enable the model to handle sequences of length x * original maximum pre-trained length.
            `original_max_position_embeddings` (`int`, *optional*):
                Used with 'dynamic', 'longrope' and 'llama3'.
                The original max position embeddings used during pretraining.
            `attention_factor` (`float`, *optional*):
                Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention
                computation.
                If unspecified, it defaults to value recommended by the implementation, using the `factor` field to infer the suggested value.
            `beta_fast` (`float`, *optional*):
                Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear
                ramp function. If unspecified, it defaults to 32.
            `beta_slow` (`float`, *optional*):
                Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear
                ramp function. If unspecified, it defaults to 1.
            `short_factor` (`List[float]`, *optional*):
                Only used with 'longrope'. The scaling factor to be applied to short contexts (<`original_max_position_embeddings`).
                Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2
            `long_factor` (`List[float]`, *optional*):
                Only used with 'longrope'. The scaling factor to be applied to long contexts (<`original_max_position_embeddings`).
                Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2
            `low_freq_factor` (`float`, *optional*):
                Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE
            `high_freq_factor` (`float`, *optional*):
                Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE
    num_attention_heads (`int`, *optional*, defaults to 8):
        Number of attention heads for each attention layer in the Transformer decoder.
    num_key_value_heads (`int`, *optional*):
        This is the number of key_value heads that should be used to implement Grouped Query Attention.
        If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
        `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used.
        When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group.
        For more details checkout [this paper](https://arxiv.org/pdf/2305.13245.pdf).
        If it is not specified, will default to `num_attention_heads`.
    attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
        Whether to use a bias in the query, key, value and output projection layers during self-attention.
    attention_dropout (`float`, *optional*, defaults to 0.0):
        The dropout ratio for the attention probabilities.
    mlp_bias (`bool`, *optional*, defaults to `False`):
        Whether to use a bias in up_proj, down_proj and gate_proj layers in the MLP layers.
    sliding_window (`int`, *optional*):
        Sliding window attention window size. If not specified, will default to `None`.
    keep_window_size (`int`, *optional*, defaults to 2048):
        The window size of tokens that are not dynamically masked, and dynamic masking is only performed when the sequence length exceeds this value.
    is_moe (`bool`, *optional*, defaults to `False`):
        Whether to use the Cross Domain Mixture of Experts, if `True`, the MoE will inherit the MLP to initialize.
    num_experts (`int`, *optional*, defaults to 16384):
        Number of routed experts in the model. This is only used when `is_moe=True`.
    num_experts_per_tok (`int`, *optional*, defaults to 64):
        Number of selected experts to route per-token.
    norm_topk_prob (`bool`, *optional*, defaults to `False`):
        Whether to normalize the topk probabilities.
    output_router_logits (`bool`, *optional*, defaults to `False`):
        Whether or not the router logits should be returned by the model. Enabling this will also
        allow the model to output the auxiliary loss, including load balancing loss and router z-loss.
    router_aux_loss_coef (`float`, *optional*, defaults to 0.001):
        The aux loss factor for the total loss.

```python
>>> from transformers import DogeConfig, DogeModel

>>> # Initializing a Doge-320M style configuration
>>> configuration = DogeConfig()

>>> # Initializing a model from the Doge-320M style configuration
>>> model = DogeModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```dogepast_key_valueszlayers.*.self_attn.q_projcolwisezlayers.*.self_attn.k_projzlayers.*.self_attn.v_projzlayers.*.self_attn.dt_projrowwisezlayers.*.self_attn.o_projzlayers.*.input_layernorm.weightsequence_parallelzlayers.*.input_residual.weightz(layers.*.post_attention_layernorm.weightz'layers.*.post_attention_residual.weightznorm.weightzlayers.*.mlp.gate_projzlayers.*.mlp.up_projzlayers.*.mlp.down_projzlayers.*.mlp.router_gatecolwise_repzlayers.*.mlp.down_embedrowwise_repzlayers.*.mlp.up_embed	input_idsinputs_embedshidden_statesattention_mask)embed_tokenslayersnormc                   > Xl         X l        X0l        X@l        XPl        X`l        Xpl        Xl        Xl        Xl	        Xl
        Xl        Xl        Xl        UU l        UU l        UU l        UU l        UU l        UU l        UU l        UU l        UU l        UU l        UU l        U R                  b,  SU R                  ;   a  U R                  S   U R                  S'   [3        U 5        Uc  Xl        [4        TU ]l  " SSU
0UD6  g )Ntype	rope_typetie_word_embeddings )
vocab_sizehidden_sizeintermediate_sizenum_hidden_layershidden_dropout
hidden_actinitializer_rangerms_norm_eps	use_cachemax_position_embeddings
rope_thetarope_scalingnum_attention_headsnum_key_value_headsattention_biasattention_dropoutmlp_biassliding_windowkeep_window_sizeis_moenum_expertsnum_experts_per_toknorm_topk_proboutput_router_logitsrouter_aux_loss_coefr   super__init__)selfr   r   r   r   r   r    r!   r"   r#   r   r$   r%   r&   r'   r(   r)   r*   r+   r,   r-   r.   r/   r0   r1   r2   r3   kwargs	__class__s                               c/var/www/html/shao/venv/lib/python3.13/site-packages/transformers/models/doge/configuration_doge.pyr5   DogeConfig.__init__   s   < %&!2!2,$!2("'>$$(#6 #6 ,!2 , 0&#6 ,$8!$8! (Vt7H7H-H-1->->v-FDk*t$ &':$ 	
 3	
	
    )r)   r*   r    r   r   r!   r   r.   r-   r$   r+   r1   r'   r/   r0   r   r(   r2   r"   r&   r%   r3   r,   r#   r   )i   i                  silug{Gz?gư>TFr<   g     @N   NFr>   FNr<   Fi @  @   FFgMbP?)__name__
__module____qualname____firstlineno____doc__
model_typekeys_to_ignore_at_inferencebase_model_tp_planbase_model_pp_planr5   __static_attributes____classcell__)r8   s   @r9   r   r      su   n` J#4"5#Y#Y 	$Y 	%i	
 	$Y 	*+> 	)*= 	34G 	23F 	* 	!) 		 	!) 	#M 	"=  	 !& &(9:#%568IJ!"_$56 ! $ ""7G
 G
r;   r   N)configuration_utilsr   modeling_rope_utilsr   r   __all__r   r;   r9   <module>rP      s&   . 4 9S
! S
l .r;   